Say suppose we go ahead with keeping the system as is and re-write it parallelly (essentially beginning) what distributed database would best help our use case.

Say suppose we go ahead with keeping the system as is and re-write it parallelly (essentially beginning) what distributed database would best help our use case. (Fast, distributed, ACID, transactional, no single point of failure, should have some GIS functionality)